The Cessna 172 experienced a complete loss of electrical power due to the failure of the alternator positive-terminal cable. To generate the reverse-bending fatigue loads that caused the degradation of the cable, its bending radius had to be less than the recommended standard. The investigation was unable to establish the decay time of the cable; however, it is unlikely that its deterioration began after the 22 February inspection; i.e., less than six hours before it broke. It is likely that the cable was bent beyond the established standard when the alternator was manipulated during maintenance work. The alternator and its wiring should have been examined on five occasions: during replacement of the alternator field cable and of the belt and during the last three periodic inspections. The investigation was unable to determine why the operator did not detect the wear and improper installation of the cable. It is possible that the staff who performed the periodic inspections and the work on the alternator were not aware of either the bending standards for electrical wires or the consequences of improper installation or both. The low-voltage indication of the electrical system made it possible to anticipate a loss of electrical power at least within the hour. The description of the aircraft's electrical system and the consequences of a low-voltage indication, as set out in the AOM, would have allowed the pilot to realize that the alternator was not meeting the electrical demand necessary for flight. Therefore, it can be concluded that the pilot was not familiar enough with the aircraft electrical system , as described in the AOM. If the pilot had recognized the nature of the problem, he would have been able to report it to the controller over the radio and then land in daylight before the failure. The terminal controller probably stopped monitoring the aircraft because he did not realize what was happening. Yet, there were clues that the aircraft's communication system had failed. Specifically, the pilot's last communication suggested that further communications were possible, because the intended track shortly brought the aircraft, flying at 3 000 feet asl, into St-Hubert Class D airspace. Accordingly, the controller had no reason to believe that the pilot had shut down the transponder and changed the radio frequency before entering Class D airspace, where two-way communication with the ACC and activation of the transponder were mandatory. If the Cessna 172 had descended below 2 000 feet asl to pass under the Class D airspace, the controller could at least have expected the pilot to select code 1200 for VFR flight. The aircraft, which was still visible on the radar screen, had been in St-Hubert Class D airspace for 4 minutes and 30 seconds when the terminal controller tried to reach it for the last time. As the pilot did not acknowledge and the transponder did not reply, the controller could have inferred that the aircraft had experienced a complete communication failure. The terminal controller should have informed St-Hubert tower of the situation, especially because the break in communications occurred at night when the Cessna 172 was flying less than 4 nm from the St-Hubert control zone within airspace where use of the two transceivers was mandatory. The tower controller might then have been advised of the presence of the Cessna 172. The following factors probably influenced the terminal controller's decision to treat the break in communications and the loss of transponder information as a normal situation: the loss of electrical power occurred in Class E airspace; the Cessna 172's primary target held steady on the intended track; and no manoeuvre was performed by the pilot to alert the radar system. If the pilot, who knew he was being tracked by radar, had flown a left-hand triangular pattern twice, the controller would have been advised of the situation and would have informed the appropriate control units of the circumstances. Given the sporadic nature of the primary radar returns and the absence of related data blocks on the radar screen that might have signalled an emergency situation, the airport controller would find it hard to detect the Cessna 172 on the radar screen and infer that the aircraft would land on the runway opposite to the active one. The airport controller could also expect the NORDO aircraft to perform the corresponding arrival procedure and obtain clearance before touching down. The airport controller visually scanned the runway and the runway centre line before clearing the Katana's take-off. The contrast between the lightless Cessna 172 and the nighttime background made the aircraft virtually indiscernible to the controller and the Katana's crew. It may be concluded that the airport controller accomplished his duties in accordance with established procedures and his assigned responsibilities. Because he thought that the radar was tracking the flight, the pilot of the Cessna 172 wrongly assumed that the St-Hubert airport controller knew of his presence and was providing aircraft separation and the availability of runway 24L, although he had not received any visible signal from the tower clearing him to join the final leg and land. The pilot had to ensure that his intentions were known, especially as it was night at the time and only visual acquisition made a safe landing possible. The aircraft's loss of electrical power led to multiple failures: all electrically powered systems, gauges and flight/engine instruments were rendered inoperative. The communication system, the interior and exterior lighting system and the navigation system stopped working. The fact that the loss of electrical power occurred at night made the emergency more complicated than a similar failure in daylight. The pilot, alone on board, had to control his aircraft, navigate, watch the available instruments and use a flashlight to consult the AOM, the navigation charts and the relevant checklists. The pilot's workload and stress level had to have been high and had to have affected his decision making. The pilot had to have a good knowledge of the following aircraft systems and procedures to continue the flight safely: aircraft electrical system, appropriate emergency procedures published in the AOM, ATS system, manoeuvres to alert radar stations, complete communication failure procedures, NORDO aircraft arrival procedures, NORDO aircraft landing procedures and the meaning of the authorized visual signals used by the tower. Although understanding these systems and procedures is mandatory to obtain a private pilot licence, the pilot's inexperience probably led him not to realize the imminence of the loss of electrical power and to overestimate the assistance of ATS. As the Cessna 172 landed midway on the runway, almost two minutes had to have elapsed between the aircraft's last recorded position 2 nm on the final leg and landing. It may be concluded that runway 24L was clear when the Cessna 172 was on the final leg, and that the Cessna 172 landed shortly after the Katana's pilot entered the runway and began his take-off run. Because the collision occurred less than 1 000 feet from the threshold of runway 06R and 42 seconds after the Katana's pilot acknowledged the take-off clearance, the length of the head-on collision course is likely to have left the pilot of the Cessna 172 little time to see the other aircraft, realize the imminence of the danger and try evasive action. If the pilot had followed the NORDO aircraft arrival procedure, he would have been able to see the traffic and identify the runway in use. It is also possible that the controller would have noticed the aircraft on the radar, reported its presence to other pilots and cleared the landing using visual signals.Analysis The Cessna 172 experienced a complete loss of electrical power due to the failure of the alternator positive-terminal cable. To generate the reverse-bending fatigue loads that caused the degradation of the cable, its bending radius had to be less than the recommended standard. The investigation was unable to establish the decay time of the cable; however, it is unlikely that its deterioration began after the 22 February inspection; i.e., less than six hours before it broke. It is likely that the cable was bent beyond the established standard when the alternator was manipulated during maintenance work. The alternator and its wiring should have been examined on five occasions: during replacement of the alternator field cable and of the belt and during the last three periodic inspections. The investigation was unable to determine why the operator did not detect the wear and improper installation of the cable. It is possible that the staff who performed the periodic inspections and the work on the alternator were not aware of either the bending standards for electrical wires or the consequences of improper installation or both. The low-voltage indication of the electrical system made it possible to anticipate a loss of electrical power at least within the hour. The description of the aircraft's electrical system and the consequences of a low-voltage indication, as set out in the AOM, would have allowed the pilot to realize that the alternator was not meeting the electrical demand necessary for flight. Therefore, it can be concluded that the pilot was not familiar enough with the aircraft electrical system , as described in the AOM. If the pilot had recognized the nature of the problem, he would have been able to report it to the controller over the radio and then land in daylight before the failure. The terminal controller probably stopped monitoring the aircraft because he did not realize what was happening. Yet, there were clues that the aircraft's communication system had failed. Specifically, the pilot's last communication suggested that further communications were possible, because the intended track shortly brought the aircraft, flying at 3 000 feet asl, into St-Hubert Class D airspace. Accordingly, the controller had no reason to believe that the pilot had shut down the transponder and changed the radio frequency before entering Class D airspace, where two-way communication with the ACC and activation of the transponder were mandatory. If the Cessna 172 had descended below 2 000 feet asl to pass under the Class D airspace, the controller could at least have expected the pilot to select code 1200 for VFR flight. The aircraft, which was still visible on the radar screen, had been in St-Hubert Class D airspace for 4 minutes and 30 seconds when the terminal controller tried to reach it for the last time. As the pilot did not acknowledge and the transponder did not reply, the controller could have inferred that the aircraft had experienced a complete communication failure. The terminal controller should have informed St-Hubert tower of the situation, especially because the break in communications occurred at night when the Cessna 172 was flying less than 4 nm from the St-Hubert control zone within airspace where use of the two transceivers was mandatory. The tower controller might then have been advised of the presence of the Cessna 172. The following factors probably influenced the terminal controller's decision to treat the break in communications and the loss of transponder information as a normal situation: the loss of electrical power occurred in Class E airspace; the Cessna 172's primary target held steady on the intended track; and no manoeuvre was performed by the pilot to alert the radar system. If the pilot, who knew he was being tracked by radar, had flown a left-hand triangular pattern twice, the controller would have been advised of the situation and would have informed the appropriate control units of the circumstances. Given the sporadic nature of the primary radar returns and the absence of related data blocks on the radar screen that might have signalled an emergency situation, the airport controller would find it hard to detect the Cessna 172 on the radar screen and infer that the aircraft would land on the runway opposite to the active one. The airport controller could also expect the NORDO aircraft to perform the corresponding arrival procedure and obtain clearance before touching down. The airport controller visually scanned the runway and the runway centre line before clearing the Katana's take-off. The contrast between the lightless Cessna 172 and the nighttime background made the aircraft virtually indiscernible to the controller and the Katana's crew. It may be concluded that the airport controller accomplished his duties in accordance with established procedures and his assigned responsibilities. Because he thought that the radar was tracking the flight, the pilot of the Cessna 172 wrongly assumed that the St-Hubert airport controller knew of his presence and was providing aircraft separation and the availability of runway 24L, although he had not received any visible signal from the tower clearing him to join the final leg and land. The pilot had to ensure that his intentions were known, especially as it was night at the time and only visual acquisition made a safe landing possible. The aircraft's loss of electrical power led to multiple failures: all electrically powered systems, gauges and flight/engine instruments were rendered inoperative. The communication system, the interior and exterior lighting system and the navigation system stopped working. The fact that the loss of electrical power occurred at night made the emergency more complicated than a similar failure in daylight. The pilot, alone on board, had to control his aircraft, navigate, watch the available instruments and use a flashlight to consult the AOM, the navigation charts and the relevant checklists. The pilot's workload and stress level had to have been high and had to have affected his decision making. The pilot had to have a good knowledge of the following aircraft systems and procedures to continue the flight safely: aircraft electrical system, appropriate emergency procedures published in the AOM, ATS system, manoeuvres to alert radar stations, complete communication failure procedures, NORDO aircraft arrival procedures, NORDO aircraft landing procedures and the meaning of the authorized visual signals used by the tower. Although understanding these systems and procedures is mandatory to obtain a private pilot licence, the pilot's inexperience probably led him not to realize the imminence of the loss of electrical power and to overestimate the assistance of ATS. As the Cessna 172 landed midway on the runway, almost two minutes had to have elapsed between the aircraft's last recorded position 2 nm on the final leg and landing. It may be concluded that runway 24L was clear when the Cessna 172 was on the final leg, and that the Cessna 172 landed shortly after the Katana's pilot entered the runway and began his take-off run. Because the collision occurred less than 1 000 feet from the threshold of runway 06R and 42 seconds after the Katana's pilot acknowledged the take-off clearance, the length of the head-on collision course is likely to have left the pilot of the Cessna 172 little time to see the other aircraft, realize the imminence of the danger and try evasive action. If the pilot had followed the NORDO aircraft arrival procedure, he would have been able to see the traffic and identify the runway in use. It is also possible that the controller would have noticed the aircraft on the radar, reported its presence to other pilots and cleared the landing using visual signals. The Cessna 172 experienced a complete loss of electrical power at night while in Class E airspace. The loss of power was caused by the failure of the alternator positive-terminal cable. The installation of the cable was not in compliance with the recommended standard and caused reverse-bending fatigue loads causing degradation. The operator did not detect the wear and improper installation of the cable in the last periodic inspection, six hours before the flight. The pilot of the Cessna 172 had observed a low-voltage indication before the failure. The pilot of the Cessna 172 was not very familiar with the consequences of low voltage. The pilot of the Cessna 172 did not follow the aircraft manufacturer's recommended procedure for low voltage. The Montreal terminal controller and the St-Hubert airport controller did not know that the Cessna 172 had experienced a complete loss of electrical power, and that the pilot would touch down on runway 24L. The terminal controller should have informed the St-Hubert tower of the presence of the Cessna 172 which was 4 nm from the St-Hubert control zone after he tried unsuccessfully to communicate with the aircraft in Class D airspace. The airport controller accomplished his duties in accordance with established procedures and his assigned responsibilities. The contrast between the lightless Cessna 172 and the nighttime background did not allow the airport controller and the Katana's crew to see the Cessna 172 on approach and on the runway. The pilot of the Cessna 172 did not follow the NORDO aircraft arrival procedure at St-Hubert Airport.Findings The Cessna 172 experienced a complete loss of electrical power at night while in Class E airspace. The loss of power was caused by the failure of the alternator positive-terminal cable. The installation of the cable was not in compliance with the recommended standard and caused reverse-bending fatigue loads causing degradation. The operator did not detect the wear and improper installation of the cable in the last periodic inspection, six hours before the flight. The pilot of the Cessna 172 had observed a low-voltage indication before the failure. The pilot of the Cessna 172 was not very familiar with the consequences of low voltage. The pilot of the Cessna 172 did not follow the aircraft manufacturer's recommended procedure for low voltage. The Montreal terminal controller and the St-Hubert airport controller did not know that the Cessna 172 had experienced a complete loss of electrical power, and that the pilot would touch down on runway 24L. The terminal controller should have informed the St-Hubert tower of the presence of the Cessna 172 which was 4 nm from the St-Hubert control zone after he tried unsuccessfully to communicate with the aircraft in Class D airspace. The airport controller accomplished his duties in accordance with established procedures and his assigned responsibilities. The contrast between the lightless Cessna 172 and the nighttime background did not allow the airport controller and the Katana's crew to see the Cessna 172 on approach and on the runway. The pilot of the Cessna 172 did not follow the NORDO aircraft arrival procedure at St-Hubert Airport. The pilot of the Cessna 172 did not follow the NORDO aircraft arrival procedure and did not make an adequate visual check before landing in the opposite direction on runway 06R just as the Katana was cleared to take off from runway 06R. The following factors contributed to the accident: the installation of the alternator positive-terminal cable was not in compliance with the recommended standard; the operator did not detect the wear and improper installation of the cable; the pilot of the Cessna 172 was not familiar with the consequences of a low-voltage indication; and the terminal controller's lack of vigilance.Causes and Contributing Factors The pilot of the Cessna 172 did not follow the NORDO aircraft arrival procedure and did not make an adequate visual check before landing in the opposite direction on runway 06R just as the Katana was cleared to take off from runway 06R. The following factors contributed to the accident: the installation of the alternator positive-terminal cable was not in compliance with the recommended standard; the operator did not detect the wear and improper installation of the cable; the pilot of the Cessna 172 was not familiar with the consequences of a low-voltage indication; and the terminal controller's lack of vigilance.